Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
1. A method for automatically comparing organ tissue properties of a patient with corresponding organ tissue properties of a group of healthy subjects, the method comprising the following steps: determining a population norm for the organ tissue properties by: a) selecting at least two different tissue properties of the organ to be investigated; b) determining for each tissue property previously selected and for each subject of said group a quantitative tissue property map; c) for each subject of the group, calculating a joint histogram from all the quantitative tissue property maps obtained for said subject; d) normalizing each previously calculated joint histogram; e) determining an averaged joint histogram by averaging the normalized joint histograms across all subjects of the healthy group, wherein said averaged joint histogram defines said population norm; automatically performing a comparison of the averaged joint histogram with a patient joint histogram obtained for the organ tissue properties of the patient, wherein the patient joint histogram is obtained through the following steps: b1) determining for each tissue property selected in step a) a quantitative tissue property map for the patient organ; and c1) calculating a joint histogram from all the quantitative tissue property maps obtained for the organ of said patient; d1) normalizing the previously calculated joint histogram; and wherein the comparison is obtained by calculating a statistical deviation of values of the patient joint histogram with respect to values of the averaged joint histogram and mapping the statistical deviation to the patient organ.
Medical imaging analysis. This invention addresses the challenge of objectively assessing organ tissue health by comparing a patient's tissue characteristics to a reference standard derived from healthy individuals. The method involves first establishing a population norm for specific organ tissue properties. This is achieved by selecting at least two distinct tissue properties of the organ of interest. For each selected tissue property, quantitative maps are generated for a group of healthy subjects. A joint histogram is then computed for each healthy subject by combining all their quantitative tissue property maps. These individual joint histograms are normalized. An averaged joint histogram, representing the population norm, is calculated by averaging the normalized joint histograms from all healthy subjects. Subsequently, a comparison is automatically performed between this population norm (the averaged joint histogram) and a patient's tissue properties. For the patient, quantitative tissue property maps are generated for each of the selected tissue properties. A joint histogram is then calculated for the patient by combining their quantitative tissue property maps. This patient joint histogram is also normalized. The comparison is made by calculating the statistical deviation between the values in the patient's joint histogram and the values in the averaged joint histogram. This statistical deviation is then mapped back to the patient's organ, providing a visual representation of how the patient's tissue properties deviate from the healthy population.
2. The method according to claim 1 , wherein calculating the joint histogram comprises: determining for each subject of said group and for each quantitative tissue property map obtained for said subject a total volume of said organ; dividing said total volume into voxels wherein each voxel is associated to a quantitative value for the tissue property of the quantitative tissue property map; dividing the quantitative values of each of the quantitative tissue property maps into bins; for each bin, counting the number of voxels lying within said bin in order to create the joint histogram.
This invention relates to medical imaging and quantitative tissue analysis, specifically for generating joint histograms from multiple quantitative tissue property maps of an organ across a group of subjects. The problem addressed is the need for a standardized method to analyze and compare tissue properties across different subjects in a consistent and quantifiable manner. The method involves calculating a joint histogram for a group of subjects, where each subject has undergone imaging to produce one or more quantitative tissue property maps of an organ. For each subject, the total volume of the organ is determined and divided into voxels, with each voxel assigned a quantitative value corresponding to the tissue property from the map. The quantitative values from all subjects' maps are then divided into bins. For each bin, the number of voxels across all subjects that fall within that bin is counted, resulting in a joint histogram. This histogram provides a statistical distribution of the tissue property values across the entire group, enabling comparative analysis. The method ensures that the joint histogram accurately reflects the distribution of tissue properties by accounting for the total organ volume and voxel-based quantification, allowing for robust statistical analysis in medical imaging studies.
3. The method according to claim 2 , wherein the step of normalizing each previously calculated joint histogram comprises dividing each bin of the joint histogram by the total number of voxels comprised within the total volume of said organ.
This invention relates to medical imaging and image processing, specifically for analyzing and normalizing joint histograms derived from volumetric medical images of an organ. The problem addressed is the variability in joint histogram data due to differences in organ size and voxel count, which can hinder accurate comparison and analysis of image features across different patients or imaging sessions. The method involves processing a joint histogram, which is a statistical representation of the co-occurrence of intensity values between two different imaging modalities or sequences, such as CT and MRI, within a defined organ volume. The joint histogram is calculated by analyzing the intensity values of corresponding voxels in the two imaging modalities and counting their occurrences. To ensure consistency and enable meaningful comparisons, the joint histogram is normalized by dividing each bin of the histogram by the total number of voxels within the entire volume of the organ. This normalization step adjusts the histogram values to account for variations in organ size and voxel count, providing a standardized representation of the joint intensity distribution. By normalizing the joint histogram in this manner, the method allows for more accurate and reliable comparisons of image features across different patients or imaging sessions, improving the robustness of medical image analysis and diagnosis. This technique is particularly useful in applications such as multi-modal image registration, tissue characterization, and disease progression monitoring.
4. The method according to claim 1 , wherein the step of normalizing each previously calculated joint histogram comprises dividing each bin of the joint histogram by the total number of voxels comprised within the total volume of said organ.
This invention relates to medical imaging and image processing, specifically for analyzing and normalizing joint histograms derived from volumetric organ data. The problem addressed is the need to standardize joint histograms to enable accurate comparison and analysis of organ characteristics across different imaging datasets. Joint histograms are statistical representations of voxel intensity distributions in medical images, but their raw values can vary significantly due to differences in imaging parameters, organ sizes, or patient-specific factors. Without normalization, these variations make it difficult to compare or analyze the data effectively. The method involves normalizing each previously calculated joint histogram by dividing each bin of the histogram by the total number of voxels within the entire volume of the organ. This step ensures that the histogram values are scaled to a consistent range, allowing for meaningful comparisons between different organs or imaging sessions. The normalization process accounts for variations in organ size and voxel count, providing a standardized representation of the joint histogram data. This technique is particularly useful in medical imaging applications where consistent and comparable data is essential for diagnosis, treatment planning, or research. By normalizing the joint histograms, the method enables more reliable analysis of organ characteristics and improves the accuracy of subsequent image processing tasks.
5. The method according to claim 1 , wherein the calculated statistical deviation is a z-score calculated from the averaged joint histogram and the patient joint histogram, and the method comprises mapping the z-score to the patient organ.
This invention relates to medical imaging analysis, specifically a method for assessing anatomical variations in patient imaging data by comparing it to a reference dataset. The problem addressed is the need for accurate, quantitative assessment of anatomical differences between a patient's organ and a standardized reference model, which is critical for diagnosis, treatment planning, and monitoring. The method involves generating a joint histogram from a reference dataset, which represents the distribution of anatomical features in a population. A patient-specific joint histogram is then created from the patient's imaging data. The method calculates a statistical deviation, specifically a z-score, by comparing the patient's joint histogram to the averaged reference joint histogram. The z-score quantifies how many standard deviations the patient's data deviates from the reference mean, providing a normalized measure of anatomical variation. This z-score is then mapped to the patient's organ, allowing for spatial visualization of deviations across different regions. The technique enables precise identification of anatomical anomalies, aiding in clinical decision-making. The use of z-scores ensures that the comparison is statistically robust and interpretable, enhancing the reliability of the analysis. This approach is particularly useful in fields such as radiology, oncology, and surgical planning, where accurate anatomical assessment is essential.
6. The method according to claim 5 , wherein the step of mapping the z-score to the patient organ comprises: pairing all voxels in the image domain of the patient organ with a corresponding bin in the patient joint histogram domain according to voxel intensity; and assigning to each voxel the z-score from the bin it belongs to in order to build a z-score map in the image domain.
This invention relates to medical imaging and statistical analysis of patient organ data. The problem addressed is the need to accurately map statistical z-scores to specific voxels within a patient's organ image to identify abnormalities or variations in tissue properties. The method involves creating a patient-specific joint histogram that correlates voxel intensities with statistical distributions. The joint histogram is constructed by analyzing the patient's organ image and categorizing voxel intensities into bins. Each bin in the joint histogram is associated with a z-score, which quantifies how far the voxel intensity deviates from a reference or expected value. The method then maps these z-scores back to the original image domain by pairing each voxel with its corresponding bin in the joint histogram based on intensity. The z-score from the bin is assigned to the voxel, resulting in a z-score map that visually represents statistical deviations across the organ. This approach enables precise identification of regions with abnormal tissue properties, aiding in diagnosis and treatment planning. The method ensures that statistical analysis is directly linked to spatial information in the image, improving accuracy and interpretability.
7. An imaging system configured for automatically comparing organ tissue properties of a patient with corresponding organ tissue properties of a group of healthy subjects, the imaging system comprising: a processor and an accessible memory, the imaging system being configured to: determine a population norm for said organ tissue properties by: a) selecting at least two different tissue properties of the organ to be investigated; b) determining for each tissue property previously selected and for each subject of said group a quantitative tissue property map; c) for each subject of the group, calculating a joint histogram from all the quantitative tissue property maps obtained for said subject; d) normalizing each previously calculated joint histogram; and e) determining an averaged joint histogram by averaging the normalized joint histograms across all subjects of the healthy group, wherein said averaged joint histogram defines said population norm; automatically perform a comparison of the averaged joint histogram with a patient joint histogram obtained for the organ tissue properties of the patient, wherein, for determining the patient joint histogram, the imaging system is configured for: b1) determining for each tissue property selected in step a) a quantitative tissue property map for the patient organ; and c1) calculating a joint histogram from all the quantitative tissue property maps obtained for the organ of said patient; d1) normalizing the previously calculated joint histogram; and and wherein the comparison is obtained by calculating a statistical deviation of values of the patient joint histogram with respect to values of the averaged joint histogram and mapping said statistical deviation to the patient organ.
This invention relates to an imaging system for comparing organ tissue properties of a patient with those of a healthy population. The system addresses the challenge of detecting abnormalities in organ tissue by leveraging statistical analysis of multiple tissue properties. The system uses a processor and memory to analyze at least two different tissue properties of an organ, such as density, elasticity, or other quantitative metrics. For a group of healthy subjects, the system generates quantitative tissue property maps for each selected property, computes a joint histogram for each subject by combining these maps, normalizes the histograms, and averages them to establish a population norm. For a patient, the system similarly generates tissue property maps, computes a joint histogram, and normalizes it. The patient's joint histogram is then compared to the population norm by calculating statistical deviations, which are mapped back to the patient's organ to highlight areas of concern. This approach enables automated detection of tissue abnormalities by identifying deviations from healthy tissue distributions.
8. The imaging system of claim 7 , configured for calculating the joint histogram by: determining for each subject of said group and for each quantitative tissue property map obtained for said subject a total volume of said organ; dividing said total volume into voxels wherein each voxel is associated to a quantitative value for the tissue property of the quantitative tissue property map; dividing the quantitative values of each of the quantitative tissue property maps into bins; for each bin, counting the number of voxels lying within said bin in order to create the joint histogram.
This invention relates to an imaging system for analyzing tissue properties in a group of subjects. The system addresses the challenge of quantifying and comparing tissue characteristics across multiple subjects by generating a joint histogram that captures the distribution of tissue properties within an organ. The system first obtains quantitative tissue property maps for each subject, which represent spatial variations in tissue properties such as density, elasticity, or perfusion. For each subject, the system calculates the total volume of the organ of interest and divides this volume into voxels, where each voxel is assigned a quantitative value from the tissue property map. The system then divides the quantitative values of each tissue property map into discrete bins. For each bin, the system counts the number of voxels that fall within that bin, thereby constructing a joint histogram. This histogram provides a statistical representation of how tissue properties are distributed across the organ and can be used to compare tissue characteristics between subjects or identify patterns in tissue property distributions. The system enables quantitative analysis of tissue properties in a standardized manner, facilitating medical research and clinical decision-making.
9. The imaging system of claim 7 , configured for dividing each bin of the joint histogram by a total number of voxels comprised within the total volume of said organ in order to normalize each previously calculated joint histogram.
This invention relates to medical imaging systems designed to analyze organ structures by generating and processing joint histograms. The system addresses the challenge of accurately quantifying and comparing tissue characteristics within an organ by normalizing joint histograms to account for variations in organ size and voxel distribution. The imaging system first constructs a joint histogram representing the statistical distribution of voxel intensities and their spatial relationships within the organ. This histogram is then normalized by dividing each bin by the total number of voxels in the organ's volume. This normalization step ensures that the histogram values are independent of the organ's size, allowing for consistent comparisons across different patients or imaging conditions. The system may also include preprocessing steps such as segmenting the organ from surrounding tissues and generating a 3D volume representation. The normalized joint histogram can then be used for further analysis, such as identifying tissue abnormalities or assessing organ health. This approach improves the reliability of quantitative imaging metrics by standardizing the data to a common scale.
10. The imaging system of claim 7 , configured for calculating a z-score from the averaged joint histogram and the patient joint histogram, and mapping the z-score to the patient organ.
This invention relates to an imaging system designed to analyze and compare medical images, particularly for evaluating patient-specific anatomical structures. The system addresses the challenge of accurately assessing organ conditions by leveraging statistical analysis to highlight deviations from expected norms. The imaging system generates a joint histogram from a dataset of reference images, representing the expected distribution of pixel intensities or other features across a population. It then creates a patient-specific joint histogram from the patient's medical images. These histograms are averaged to establish a baseline reference model. The system calculates a z-score by comparing the patient's joint histogram to the averaged reference histogram, quantifying how much the patient's data deviates from the norm. This z-score is then mapped to the patient's organ, providing a spatial representation of areas where the patient's anatomy differs significantly from the reference population. The system may also include preprocessing steps to align or normalize the images before analysis, ensuring accurate comparisons. The z-score mapping helps clinicians identify abnormal regions in the patient's organ, aiding in diagnosis or treatment planning.
11. A non-transitory computer-readable medium encoded with executable instructions that, when executed, cause one or more imaging systems to perform the method according to claim 1 .
This invention relates to medical imaging systems, specifically addressing the challenge of accurately detecting and analyzing anatomical structures in medical images. The system uses computer vision and machine learning techniques to process imaging data from modalities such as MRI, CT, or ultrasound. The method involves capturing raw imaging data, pre-processing it to enhance image quality, and applying segmentation algorithms to identify and isolate anatomical regions of interest. These regions are then analyzed for abnormalities, such as tumors or lesions, using pattern recognition and classification models. The system also includes a user interface for visualizing results and allowing clinicians to interact with the analysis. The method further incorporates adaptive learning, where the system improves its accuracy over time by incorporating feedback from user corrections or additional training data. The invention aims to reduce diagnostic errors, improve efficiency in medical imaging workflows, and provide more reliable detection of pathological conditions. The system is designed to integrate with existing imaging hardware and software platforms, ensuring compatibility with current clinical environments.
Unknown
May 19, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.